Some Simple Guidelines for Effective Data Management

Abstract

Data are at the heart of empirically based sciences, serving as the primary evidence supporting (or refuting) models of the way that our natural world operates. While most scientists spend much time thinking about the types of data they need to further their studies, and often collect and analyze their data using sophisticated techniques and approaches, relatively little effort is spent considering how to store their data. An unfortunate result of this cursory data management is that much valuable information gets lost. This is a waste when one recognizes that scientifi c observations capture the values and relationships among an interesting set of objects or processes at some place and time, and typically contain information useful for investigating research questions that were never anticipated when the data were originally collected. Many ecological studies can benefi t from access to additional data with similar thematic foci, or collected from the same spatial region during the same time period. Synthetic, integrative research in ecology in particular depends on using data typically collected by multiple parties that were not coordinated in their research approach. For these reasons, it is increasingly important to store and document scientifi c data in ways that facilitate their effective retrieval and interpretation in the future. This document provides some simple guidelines for effective data management, which, if put into practice, will benefi t the original data owner as well as enhance prospects for the long-term preservation and re-use of the data by other researchers. The rules of thumb presented here facilitate rapid and accurate interpretation of the data, and pre-dispose the data to more effective processing by computers. We hope that you will embrace these simple rules, which are based on general data-modeling principles that are broadly relevant to any types of information that you might collect.